Update for Gradio 6

app.py

@@ -180,16 +180,43 @@ def run_fewshot_classifier(old_revision: str, new_revision: str):
     return run_classifier(old_revision, new_revision, prompt_style="few-shot")
 
 
+def compute_confidence(
+    heuristic_noteworthy,
+    fewshot_noteworthy,
+    judge_noteworthy,
+    heuristic_rationale,
+    fewshot_rationale,
+    judge_reasoning,
+):
+    """
+    Compute a confidence label using the noteworthy booleans.
+    """
+    # Return None if any of the rationales or reasoning is missing.
+    if not heuristic_rationale or not fewshot_rationale or not judge_reasoning:
+        return None
+    if heuristic_noteworthy == fewshot_noteworthy == judge_noteworthy:
+        # Classifiers and judge all agree
+        return "High"
+    elif heuristic_noteworthy != fewshot_noteworthy:
+        # Classifiers disagree, judge decides
+        return "Moderate"
+    else:
+        # Classifiers agree, judge vetoes
+        return "Questionable"
+
+
 @logfire.instrument("Step 4: Run judge")
 def run_judge(
     old_revision: str,
     new_revision: str,
-    heuristic_rationale,
-    fewshot_rationale,
+    heuristic_noteworthy: bool,
+    fewshot_noteworthy: bool,
+    heuristic_rationale: str,
+    fewshot_rationale: str,
     judge_mode: str,
 ):
     """
-    Run classification models and judge on the revisions.
+    Run judge on the revisions and classifiers' rationales.
 
     Args:
         old_revision: Old revision text
@@ -199,18 +226,18 @@ def run_judge(
         judge_mode: Mode for judge function ("unaligned", "aligned-fewshot", "aligned-heuristic")
 
     Returns:
-        Tuple of (noteworthy, reasoning) (bool, str)
+        Tuple of (noteworthy, noteworthy_text, reasoning, confidence) (bool, str, str, str)
     """
 
     # Values to return if there is an error
-    noteworthy, reasoning = None, None
+    noteworthy, noteworthy_text, reasoning, confidence = None, None, None, None
     if (
         not old_revision
         or not new_revision
         or not heuristic_rationale
         or not fewshot_rationale
     ):
-        return noteworthy, reasoning
+        return noteworthy, noteworthy_text, reasoning, confidence
 
     try:
         # Run judge
@@ -232,52 +259,27 @@ def run_judge(
         error_msg = f"Error running judge: {str(e)}"
         raise gr.Error(error_msg, print_exception=False)
 
-    return noteworthy, reasoning
-
-
-def format_noteworthy(noteworthy, reasoning):
-    """
-    Format judge's noteworthy label as text.
-    """
+    # Format noteworthy label (boolean) as text
     if not reasoning:
-        # If the reasoning is empty, return nothing
-        return None
-    else:
-        # Format noteworthy boolean as text
-        return str(noteworthy)
-
-
-def compute_confidence(
-    heuristic_noteworthy,
-    fewshot_noteworthy,
-    judge_noteworthy,
-    heuristic_rationale,
-    fewshot_rationale,
-    judge_reasoning,
-):
-    """
-    Compute a confidence label using the noteworthy booleans.
-    """
-    # Return None if any of the rationales or reasoning is missing.
-    if not heuristic_rationale or not fewshot_rationale or not judge_reasoning:
-        return None
-    if heuristic_noteworthy == fewshot_noteworthy == judge_noteworthy:
-        # Classifiers and judge all agree
-        return "High"
-    elif heuristic_noteworthy != fewshot_noteworthy:
-        # Classifiers disagree, judge decides
-        return "Moderate"
+        noteworthy_text = None
     else:
-        # Classifiers agree, judge vetoes
-        return "Questionable"
+        noteworthy_text = str(noteworthy)
+
+    # Get confidence score
+    confidence = compute_confidence(
+        heuristic_noteworthy,
+        fewshot_noteworthy,
+        noteworthy,
+        heuristic_rationale,
+        fewshot_rationale,
+        reasoning,
+    )
 
+    return noteworthy, noteworthy_text, reasoning, confidence
 
-# Setup theme without background image
-theme = gr.Theme.from_hub("NoCrypt/miku")
-theme.set(body_background_fill="#FFFFFF", body_background_fill_dark="#000000")
 
 # Create Gradio interface
-with gr.Blocks(theme=theme, title="Noteworthy Differences") as demo:
+with gr.Blocks(title="Noteworthy Differences") as demo:
     with gr.Row():
         with gr.Column(scale=2):
             gr.Markdown(
@@ -360,12 +362,12 @@ with gr.Blocks(theme=theme, title="Noteworthy Differences") as demo:
                     lines=1,
                     interactive=False,
                 )
-            rerun_btn = gr.Button("Rerun Model", variant="primary")
+            rerun_btn = gr.Button("Rerun Model")
 
-    # Hidden checkboxes to store boolean values
-    heuristic_noteworthy = gr.Checkbox(visible=False)
-    fewshot_noteworthy = gr.Checkbox(visible=False)
-    judge_noteworthy = gr.Checkbox(visible=False)
+    # States to store boolean values
+    heuristic_noteworthy = gr.State()
+    fewshot_noteworthy = gr.State()
+    judge_noteworthy = gr.State()
 
     random_btn.click(
         fn=get_random_wikipedia_title,
@@ -407,28 +409,13 @@ with gr.Blocks(theme=theme, title="Noteworthy Differences") as demo:
         inputs=[
             old_revision,
             new_revision,
-            heuristic_rationale,
-            fewshot_rationale,
-            judge_mode_dropdown,
-        ],
-        outputs=[judge_noteworthy, judge_reasoning],
-        api_name=False,
-    ).then(
-        fn=format_noteworthy,
-        inputs=[judge_noteworthy, judge_reasoning],
-        outputs=[noteworthy_text],
-        api_name=False,
-    ).then(
-        fn=compute_confidence,
-        inputs=[
             heuristic_noteworthy,
             fewshot_noteworthy,
-            judge_noteworthy,
             heuristic_rationale,
             fewshot_rationale,
-            judge_reasoning,
+            judge_mode_dropdown,
         ],
-        outputs=[confidence],
+        outputs=[judge_noteworthy, noteworthy_text, judge_reasoning, confidence],
         api_name=False,
     )
 
@@ -449,30 +436,20 @@ with gr.Blocks(theme=theme, title="Noteworthy Differences") as demo:
         inputs=[
             old_revision,
             new_revision,
-            heuristic_rationale,
-            fewshot_rationale,
-            judge_mode_dropdown,
-        ],
-        outputs=[judge_noteworthy, judge_reasoning],
-        api_name=False,
-    ).then(
-        fn=format_noteworthy,
-        inputs=[judge_noteworthy, judge_reasoning],
-        outputs=[noteworthy_text],
-        api_name=False,
-    ).then(
-        fn=compute_confidence,
-        inputs=[
             heuristic_noteworthy,
             fewshot_noteworthy,
-            judge_noteworthy,
             heuristic_rationale,
             fewshot_rationale,
-            judge_reasoning,
+            judge_mode_dropdown,
         ],
-        outputs=[confidence],
+        outputs=[judge_noteworthy, noteworthy_text, judge_reasoning, confidence],
         api_name=False,
     )
 
 if __name__ == "__main__":
-    demo.launch()
+
+    # Setup theme without background image
+    theme = gr.Theme.from_hub("NoCrypt/miku")
+    theme.set(body_background_fill="#FFFFFF", body_background_fill_dark="#000000")
+
+    demo.launch(theme=theme)

requirements.txt

@@ -2,7 +2,7 @@ google-genai
 pydantic
 pandas
 dotenv
-gradio
+gradio>=6.0.1
 requests
 logfire
 opentelemetry-instrumentation-google-genai